Claude Opus fast mode is now available in research preview on the API and in Claude Code. The feature makes Claude Opus 2.5 times faster at a higher cost per token, allowing developers to toggle the capability on for time-sensitive work and off when cost efficiency takes priority.
Fast mode uses Claude Opus with a different API configuration that prioritizes speed over cost efficiency. According to Anthropic, users get identical quality and capabilities, just with faster responses. The feature works on both Opus 4.6 and Opus 4.7 but is not available on Sonnet, Haiku, or other models.
How to Use Claude Opus Fast Mode
In Claude Code, users can toggle fast mode on and off using the /fast command. This works in both the CLI and the VS Code Extension. By default, /fast runs on Opus 4.6. To run fast mode on Opus 4.7 instead, developers can set the CLAUDE_CODE_ENABLE_OPUS_4_7_FAST_MODE environment variable. The feature requires Claude Code version 2.1.36 or later. Users can check their version by running claude –version.
Pricing and Availability for Claude Opus Fast Mode
Claude Opus fast mode pricing is $30 per 150 million tokens on both Opus 4.6 and Opus 4.7. The feature is available to all Claude Code users on subscription plans, including Pro, Max, Team, and Enterprise tiers, as well as users on Claude Console. For subscription plan holders, fast mode is available through extra usage only and is not included in standard subscription rate limits.
When to Use Fast Mode
Fast mode is designed for interactive work that demands speed, such as rapid iteration during development or live debugging sessions. Developers should toggle fast mode on when latency matters more than cost and toggle it off during workflows where cost optimization is the priority.
Research Preview Status
Fast mode remains in research preview, meaning the feature, pricing, and availability may change based on user feedback. For Claude Code users, fast mode for Opus 4.7 is currently opt-in today and will become the default model for fast mode on Thursday. Anthropic continues to gather feedback from developers as it refines the offering.
The rollout reflects broader industry trends in artificial intelligence optimization, where developers increasingly demand faster inference times for software development workflows. Fast mode for Claude Opus provides developers with greater flexibility in balancing speed and cost based on their specific use case requirements.
